Meta's Muse Spark 1.2 Hits Pareto Frontier at 1/5th of Claude's Cost
ArtificialAnlys · x · 2026-08-07
According to the latest evaluation data from Artificial Analysis, Meta's Muse Spark 1.2 model performs exceptionally well on the Pareto frontier of Intelligence Index vs. Cost per Task.
The model scores only 6 points lower in intelligence compared to Claude Opus 5 but at approximately 1/6th of the cost. Priced at $1.25 per 1M input tokens and $4.25 per 1M output tokens, it delivers intelligence comparable to Claude Opus 4.8 at a fifth of the cost per task, while significantly undercutting GPT-5.6.
Related event: Meta Muse Spark 1.2 Tops Cost-Efficiency Frontier, Costing 1/6 of Claude(2 posts)→
More from Models
- OpenRouter Silently Drops Reasoning Effort Params, Skewing Model Benchmarks — PawelHuryn · 2026-08-07
- Grok 4.5 Beats Kimi K3 at 13x Lower Cost in Agent Task Test — rohanpaul_ai · 2026-08-07
- Meta's Muse Spark 1.2 Hits Pareto Frontier at 1/6th the Cost of Claude — ArtificialAnlys · 2026-08-07
- OpenAI's Upcoming Device to Focus on Personality, But Can the Model Deliver? — Angaisb_ · 2026-08-07
- Rabdos Launches Math AI Benchmark; Claude Opus 5 Takes the Lead — AI4Code · 2026-08-07
- Report: ByteDance Discussing 5-Trillion Parameter AI Model — ZeroStateReflex · 2026-08-07